Why does my ensemble still have so many repetitive (IoU > thresh I set) bboxes?

[RainbowSun11Q2H]

Thank you for your code.

However, when I use this method, I have so many repetitive bboxes. Do you have some advices?

Code is following: boxes, score12, label = example_nms_2_models(boxes_list, labels_list, scores_list, draw_image=False, method=3, iou_thr=0.5, thresh=0.1, sigma=0.5)

I just use nms method because the limitation of running time. An example of results is following: (Pdb) boxes array([[0.73385417, 0.34259259, 0.81041667, 0.45648148], [0.74930206, 0.36653722, 0.79911662, 0.45000624]]) The labels are same. Obviously, the IoU of these two boxes should be over than 0.5. Looks like NMS doen not work in ensemble. Right?

[ZFTurbo]

Can you provide more deatils? For example, 2 boxes with large IoU can be present together in case they have different labels (classes).

[ZFTurbo]

def bb_intersection_over_union(A, B):
    xA = max(A[0], B[0])
    yA = max(A[1], B[1])
    xB = min(A[2], B[2])
    yB = min(A[3], B[3])

    # compute the area of intersection rectangle
    interArea = max(0, xB - xA) * max(0, yB - yA)

    if interArea == 0:
        return 0.0

    # compute the area of both the prediction and ground-truth rectangles
    boxAArea = (A[2] - A[0]) * (A[3] - A[1])
    boxBArea = (B[2] - B[0]) * (B[3] - B[1])

    iou = interArea / float(boxAArea + boxBArea - interArea)
    return iou

iou = bb_intersection_over_union(
        (0.73385417, 0.34259259, 0.81041667, 0.45648148),
        (0.74930206, 0.36653722, 0.79911662, 0.45000624),
)
print(iou)

It gives me 0.476852 < 0.5

[shonenkov]

@ZFTurbo I have found the same problem, maybe this example can help

def bb_intersection_over_union(A, B):
    xA = max(A[0], B[0])
    yA = max(A[1], B[1])
    xB = min(A[2], B[2])
    yB = min(A[3], B[3])

    # compute the area of intersection rectangle
    interArea = max(0, xB - xA) * max(0, yB - yA)

    if interArea == 0:
        return 0.0

    # compute the area of both the prediction and ground-truth rectangles
    boxAArea = (A[2] - A[0]) * (A[3] - A[1])
    boxBArea = (B[2] - B[0]) * (B[3] - B[1])

    iou = interArea / float(boxAArea + boxBArea - interArea)
    return iou

iou = bb_intersection_over_union(
        (0.99804688, 0.58007812, 0.81835938, 0.43554688),
        (0.94726562, 0.55078125, 0.77734375, 0.4140625),
)
print(iou)
>>> 0.0

[ZFTurbo]

You have y2 less than y1 and x2 less than x1. I probably need to fix it in code.

[ZFTurbo]

I fixed problem. Earlier method works incorrect if x2 (or y2) was less than x1 (or y1), now it's automatically fixed with warning message. So latest version must be totally ok.